BIROn - Birkbeck Institutional Research Online Dewaele, Jean-Marc and MacIntyre, P. (2019) The predictive power of multicultural personality traits, learner and teacher variables on foreign language enjoyment and anxiety. In: Sato, M. and Loewen, S. (eds.) Evidence-based second language pedagogy: A collection of Instructed Second Language Acquisition studies. Abingdon, UK: Routledge. ISBN 9780815392538. Downloaded from: https://eprints.bbk.ac.uk/id/eprint/24442/ Usage Guidelines: Please refer to usage guidelines at https://eprints.bbk.ac.uk/policies.html or alternatively contact [email protected].
21
Embed
BIROn - Birkbeck Institutional Research Onlinerespond with feelings of anxiety when faced with FL demands. FL learners that suffer from FLCA “have the trait of feeling state anxiety
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
BIROn - Birkbeck Institutional Research Online
Dewaele, Jean-Marc and MacIntyre, P. (2019) The predictive power ofmulticultural personality traits, learner and teacher variables on foreignlanguage enjoyment and anxiety. In: Sato, M. and Loewen, S. (eds.)Evidence-based second language pedagogy: A collection of InstructedSecond Language Acquisition studies. Abingdon, UK: Routledge. ISBN9780815392538.
that the females’ heightened emotionality might boost the acquisition and use of the
FL, given the motivational qualities of emotion (MacIntyre, 2002). To examine
enjoyment more detail, a factor analysis identified two sub-dimensions of FLE. A
social aspect reflected in shared legends, classroom laughter, and pleasant
relationships with teachers and peers. The private side of enjoyment is reflected in
internal feelings such as pride, having fun, and a sense of accomplishment (Dewaele
& MacIntyre, 2016). Social and private dimensions of enjoyment are inter-correlated
and work together to form a cohesive feeling. The authors cautioned about risks in
focusing on only the pleasant, internal feelings associated with enjoyment in contrast
to the a more complete sense of accomplishment, growth in relationships, and
progress in shared learning. The importance of social versus private dimensions of
emotion helps focus our presentation of the learners’ descriptions of the sources of
their classroom emotions.
Dewaele and MacIntyre (2014) showed that specific positive classroom
activities that empowered students and gave them a choice in shaping an activity to
match their concerns and interests boosted levels of FLE. Funny and encouraging
teachers who praised students for good performance and deflected potential
embarrassment with sympathetic laughter also boosted learners’ FLE. In general,
having close relationships with peers also increased the potential for FLE. The
importance of climate created by people in the classroom is underscored by results
showing high levels of student engagement and teacher support are linked with lower
levels of FLCA (Palacios, 1998; Piechurska-Kuciel, 2011) and more FLE.
Relationships play a crucial role in the emotions that learners experience (Gregersen
& MacIntyre, 2014). The crucial role of classroom climate in shaping the experience
of FLE and FLCA is emerging from both quantitative (e.g., De Smet, Mettewie,
Galand, Hiligsmann & Van Mensel, to appear; Piniel and Albert, to appear) and
qualitative studies (Pavelescu & Petrić, to appear).
However, data are beginning to suggest classroom climate might affect FLE and
FLA in different ways. Results of a study of 189 British learners suggest that,
compared to FLE, FLCA may be less related to the teacher and teaching practices
(Dewaele et al., 2017). Using a pseudo-longitudinal design, Dewaele and Dewaele
(2017) found FLE and FLCA evolved over time differently and were influenced by
different learner and teacher variables. The authors emphasize that sources of positive
and negative emotions change as learners become older. Novel methodologies, using
an individual-level, dynamic (idiodynamic) approach to examine relationships
between FLE and FLA are showing sharp, short term fluctuation in both FLE and
FLCA, and a relationship that appears to be highly complex (Boudreau, MacIntyre, &
Dewaele, to appear). At times, FLA and FLE appear to be coordinate and at other
times they appear to operate independently; there is a wide range of correlations
observed between anxiety and enjoyment on a moment-by-moment basis, suggesting
that research is needed to better understand why learners can tell us that “even when
I’m feeling a little anxious about it, I still found it fun.” Although such a statement
might seem internally inconsistent, it never-the-less matches learners’ conflicted and
ambivalent emotional experiences (Boudreau et al., to appear; MacIntyre, 2007). The nascent literature is suggesting that contrasting positive and negative
emotions in the same study provides an interesting avenue for better understanding
both commonalities and differences between emotions. In the present study, we will
examine that relationship and the predictive value offered by learner-internal and
contextual variables for both emotions. Specifically, we examine a collection of 19
variables, including demographic, language-related, teacher-related factors, and
learner personality as they correlate with FLE and FLCA. The following research questions were explored in the study:
RQ1: What is the relationship between FLE and FLCA?
RQ2: Is there a difference in FLE and FLCA of female and male participants?
RQ3: Do FLE and FLCA vary in function of the geographical area where the
FL is studied?
RQ4: What are the cumulated effects of age, number of languages known,
learner-internal and teacher-related variables and multicultural personality
traits on learners’ FLE & FLCA?
RQ5: What sources of FLE and FLCA can be identified in participants’
accounts of enjoyable and anxiety-provoking episodes in their FL classes?
METHODS
Participants
The study attracted 750 participants (533 females, 207 malesi, mean age: 26, SD =
10). A large majority of participants studied in Europe. Close to half of participants
reported studying English as a FL (n = 354), followed by French, Spanish, German,
Japanese. Participants (n = 630) reported the result on their last FL test: ranging from
30% to 100%, with a mean of 81.5% (SD = 12.3). Asked about the frequency of use
of the FL outside school (ranging from ‘very rarely’ = 1 to ‘very frequently’ = 5),
participants reported occasional general use outside school (Mean = 2.8, SD = 1.0)
and more frequent of the FL for reading (Mean = 3.3, SD = 1.1).
Participants also reported how positive their attitude was toward the FL they
were studying, ranging from ‘very unfavourable’ to ‘very favourable’ (Mean = 4.4,
SD = 0.8). They were also asked to indicate on a 5-point Likert scale what their
attitude was toward their FL teacher (ranging from ‘very negative’ to ‘very positive’)
which yielded a Mean score of 4.1, SD = 1.0. A further number of items with 5-point
Likert scales ranging from ‘not at all’ to ‘very much so’ enquired about various
aspects of the teacher, including how strict the teacher was (Mean = 2.4, SD = 1.0),
how friendly (Mean = 4.2, SD = 0.9), how much FL the teacher used in class (Mean =
4.4, SD = 0.9), how much the teacher joked (Mean = 3.2, SD = 0.9) and how
predictable the teacher was (Mean = 3.5, SD = 0.8). Other demographic information
can be found in the Appendix.
Instruments
In addition to the sociobiographical and language-related items in Appendix A,
participants completed the following measures:
1. FLE: Ten items were extracted from the Foreign Language Enjoyment (FLE)
questionnaire (Dewaele & MacIntyre, 2014). Items reflect both the social and
private dimensions of FLE all items were positively phrased Internal
reliability was good (Cronbach alpha = .89; Mean = 3.9, SD = 0.6).
2. FLA: Eight items which reflected physical symptoms of anxiety, nervousness
and lack of confidence were extracted from the FLCAS (Horwitz et al., 1986).
Two FLCA items were phrased to indicate low anxiety (reverse scored) and
six were phrased to indicate high anxiety. Internal reliability was good
(Cronbach alpha = .87, mean = 2.4, SD = 0.8).
3. Multicultural Personality Questionnaire (MPQ, short form): The MPQ (van
der Zee et al., 2013) has forty items that measure five personality dimensions
relevant to multicultural success.
Cultural Empathy: this measures the ability to empathize with the feelings,
thoughts and of individuals from a different cultural background. Items
include ‘Sympathizes with others’ and ‘Sets others at ease’. Cronbach’s α:
.81.
Open-mindedness: this refers to an open and unprejudiced attitude towards
outgroup members and towards different cultural norms and values. Items
include ‘Likes to imagine solutions for problems’ and ‘Has a broad range
of interests’. Cronbach’s α: .76.
Social Initiative: this is defined as a tendency to approach social situations
in an active way and to take initiative. Items include ‘Leaves the initiative
to others to make contacts’ and ‘Takes the lead’. Cronbach’s α: .80.
Emotional Stability: this dimension refers to a tendency to remain calm in
stressful situations versus a tendency to show strong emotional reactions
under stressful circumstances. Items include ‘Is nervous’ and ‘Keeps calm
when things don’t go well’. Cronbach’s α: .76.
Flexibility: the final dimension is described as the ability to learn from
experiences. Elements of flexibility, such as the ability to learn from
mistakes and adjustment of behaviour, whenever it is required, are
associated with the ability to learn from new experiences in particular.
Items include ‘Works according to plan’ and ‘Looks for regularity in life’.
Cronbach’s α: .81.
4. Open-ended questions: The questionnaire finished with two open questions.
The first, asked “describe one specific event or episode in your foreign
language class that you really enjoyed, and describe your feeling in as much
detail as you can” (444 participants responded, 26,309 words). The second
asked “describe one specific event or episode in your foreign language class
that made you really anxious, and describe your feeling in as much detail as
you can” (463 participants, 26,384 words). Emotion-related episodes were
coded according to the persons involved. We created four categories labelled
‘self,’ ‘self-peer,’ ‘self-teacher,’ and finally ‘self-peers-teacher’. These four
broad and sometimes overlapping categories helped to explicate reasons why
the participant experienced FLE or FLCA. For reporting purposes, we chose
data extracts that were most representative of the category, most poignant, and
we found most interesting.
Following the principles of convergent parallel design (cf. Creswell & Plano
Clark, 2011, pp. 70-71), we collected the quantitative and qualitative data at the same
time, prioritized both methods equally, kept the strands separate during analysis
before mixing the results during discussion and interpretation of the findings. This
approach allowed us to triangulate the methods “by directly comparing and
contrasting quantitative results with qualitative findings for corroboration and
validation purposes” (p. 77).
The research design and questionnaire obtained approval from the Ethics
Committee of the school of Social Sciences, History and Politics at Birkbeck.
RESULTS
A look at the distribution of FLCA and FLE scores, and the calculation of Q-Q
plots (figure 1 and 2) suggests that they follow a normal distribution reasonably well
except for the extreme tail for FLCA (values below 1.5) and the region below 2.5 for
FLE. We thus opted for the more powerful parametric statistics to address RQ1
through RQ4.
Figure 1: Normal Q-Q plot of FLCA
Figure 2: Normal Q-Q plot of FLE
RQ1 tests the correlation between FLE and FLCA. A significant negative
correlation was obtained (r = -.28, p < .0001). With only 7.8% overlapping variance,
we found only a slight tendency for participants with higher scores on FLE to have
lower scores on FLCA, but the correlation is not strong.
To answer RQ2 (gender differences in FLE and FLCA), we ran an
independent t-test. The results show no difference in mean FLE scores (Mean
Females = 3.93, SD = .63, Mean Males = 3.97, SD = .55) t (738) = -.87, p = .42).
However, a highly significant difference emerged for FLCA. Female participants
showed higher mean anxiety ratings (Mean Females = 2.90, SD = .92) than males
(Mean Males = 2.58, SD = .86) t (738) = 4.25, p < .0001.
For RQ3, a one-way ANOVA showed a significant effect of geographic area
on FLE (F (5, 744) = 4.86, p < .0001, eta2 = .032. Participants studying in Australia
(Mean = 4.2, SD = .46) and North America (Mean = 4.1, SD = .58) reported the
highest mean levels of FLE, significantly higher than Africa (Mean = 3.7, SD = .77).
The other regions were in-between, with South America (Mean = 4.0, SD = .53),
Europe (Mean = 3.9, SD = .57), and Asia (Mean = 3.8, SD = .66) not significantly
different from each other or from the other regions. For FLCA, no significant effect
for geographic area was found (F (5, 744) = 2.03, p = .072).
RQ4 dealt with the relationship between the sociodemographic variables and
FLE and FLCA. As a first step, we ran Pearson correlation analyses, to reveal
significant relationships (see Table 1). Following-up on the correlations, independent
variables that were linked significantly (p < .01) with the dependent variables were
included in a stepwise linear regression analysis in order to identify the strongest
predictors of FLE and FLCA.
Table 1:
Independent variables correlations with, and regressions predicting, FLE and FLCA
Variable
r with
FLCA
r with
FLE
Beta
FLCA
Beta
FLE
Age -.089* .073*
Number of languages -.194** .039 -.13
Attitude towards FL -.189** .338** .10
FL test result -.183** .266** .11
Attitude FL teacher -.196** .478** -.07 .30
Strictness of teacher .066 -.140**
Friendliness teacher -.098** .337** .14
Teacher FL use -.018 .155**
Teacher’s predictability -.008 -.03
Teacher’s joking -.111** .334** .10
Relative Standing -.391** .244** -.23 .08
FL level -.262** .119** -.10
Frequency of Use out of school -.257** .188**
Frequency of Reading out of
school -.144** .127**
Cultural Empathy -.086* .344** .20
Flexibility -.167** -.059
Social Initiative -.407** .311** -.14 -.09
Openmindedness -.287** .316**
Emotional Stability -.528** .190** -.42
* p < .05, ** p < .01 (2-tailed)
Inter-correlations among the independent variables are presented in Table 2.
Table 2: Inter-correlations between the predictor variables